The Lung Image Database Consortium (LIDC) image collection consists of diagnostic and lung cancer screening thoracic CT scans with marked-up annotated lesions. It is a web-accessible international resource for development, training, and evaluation of computer-assisted diagnostic (CAD) methods for lung cancer detection and diagnosis. |
Collection Statistics |
|
---|---|
Modalities |
CT |
Number of Patients |
399 |
Number of Studies |
|
Number of Series |
|
Number of Images |
|
You can view and download these images on the Cancer Imaging Archive. You will need an user account to log in. Simply follow these steps:
This will return the full list of cases included in the collection. To download the associated DICOM images:
These links help describe how to use the .XML annotation files which are packaged along with the images in NBIA.
Annotation and Markup Issues/Comments
This link provides a list of available cases and the associated size of each identified nodule.
For a limited set of cases, LIDC sites were able to identify diagnostic data associated with the case. Data was collected for as many cases as possible and is associated at two levels:
At each level, data was provided as to whether the nodule was:
For each lesion, there is also information provided as to how the diagnosis was established including options such as:
You can download this Diagnosis Data at: ^LIDC Diagnosis Data-01-08-10.xls||\
Note: Data is still arriving from the LIDC sites, and so in the coming months this spreadsheet will be updated as more information is received
As part of an effort to move towards standard formats for annotation and markup a project has been undertaken to convert this data from the LIDC project into Annotated Image Markup format (AIM). AIM is a standard which was developed out of the caBIG program. A related section of the wiki has been dedicated to covering the activities of this project and can be found here: LIDC Conversion to AIM
MAX ("multi-purpose application for XML") performs nodule matching and pmap generation based on the LIDC's blinded and unblinded read responses. It also performs certain QA and QC tasks and other XML-related tasks.
MAX is written in Perl and was developed under RedHat Linux. It has been run under Windows.
Downloading MAX and its associated files implies acceptance of the following notice (also available here|^CopyrightAndLicense.txt||\ and in the distro as a text file):
DISCLAIMER: MAX is not guaranteed to process all input correctly. Possible errors include (but are not limited to) the inability to process correctly some types of nodule ambiguity (where nodule ambiguity refers to overlap between nodule markings having complicated shapes or to overlap between a nodule marking and a non-nodule mark).
Download the distro (max-V107.tgz)|^max-V107.tgz||\; view/download ^ReadMe.txt||\ (a text file that is also included in the distro).